|| || || || || ||

---

Impala::Core::Tracking::Classifier Class Reference

Implemantation of the classifier used by TrackerBackFore ("Tracking aspects of the foreground against the background" Hieu T. More...

#include <Classifier.h>

Collaboration diagram for Impala::Core::Tracking::Classifier:

[legend]List of all members.

Public Member Functions
	Classifier (int size, int dimensionslity)
virtual	~Classifier ()
void	SetVec (const Array::Array2dScalarReal64 &v, int index)
	Sets the initial value of vector[index].
void	UpdateVec (const Array::Array2dScalarReal64 &v, int index)
	Used to update the value of vector[index] by blending with the previous value.
void	GetVector (int index, Array::Array2dScalarReal64 &v)
void	GetClassifiers (int index, Array::Array2dScalarReal64 &v)
void	SetBackground (Array::Array2dScalarReal64 &background)
	analogous to SetVec
void	Update (Array::Array2dScalarReal64 &background)
	Uses the provided background to Retrain the classifier.
void	WriteSimpleDebug (std::ostream &stream)
void	WriteDebug (std::ostream &stream)
Private Member Functions
void	UpdateBackGroundModel (Array::Array2dScalarReal64 &background)
void	UpdateClassifiers ()
void	MultiplyWithTranspose (Array::Array2dScalarReal64 *&matrix, const Array::Array2dScalarReal64 *vector)
void	MultiplyWithTranspose (Array::Array2dScalarReal64 *&matrix, const double *vec)
void	AddMatrixWeighted (Array::Array2dScalarReal64 *&matrix, const Array::Array2dScalarReal64 *addmatrix, double weight)
void	MeanFromSamples (Array::Array2dScalarReal64 *&mean, const Array::Array2dScalarReal64 *samples)
void	SommateTransposeMultiplications (Array::Array2dScalarReal64 *&matrix, const Array::Array2dScalarReal64 *samples)
void	AddLambdaDiagonal (Array::Array2dScalarReal64 *&matrixDst, const Array::Array2dScalarReal64 *matrixSrc)
void	GetNormalisedObjectVector (Array::Array2dScalarReal64 *&vector, int index)
double	CrossProduct (const Array::Array2dScalarReal64 *vector1, const Array::Array2dScalarReal64 *vector2)
void	SetClassifier (int index, double k_i, const Array::Array2dScalarReal64 *correlatedVector)
Private Attributes
double	mGamma
double	mLambdaFactor
int	mSize
int	mDimensionality
Array::Array2dScalarReal64 *	mMatCovariance
Array::Array2dScalarReal64 *	mBackgroundMean
Array::Array2dScalarReal64 *	mObjectVectors
Array::Array2dScalarReal64 *	mClassifiers

---

Detailed Description

Implemantation of the classifier used by TrackerBackFore ("Tracking aspects of the foreground against the background" Hieu T.

Nguyen and Arnold Smeulders). Further comments may refer to equations from this paper.

/todo This classifier should have been moved to Core/Training. It might even be possible to fit it into the class hierachy of Core/Training, except that making a TrainDataSource for the tracking data might be a bit of a hassle.

/note This class assumes that all vectors have the same dimensionality as declared in the constructor.

/todo Some of the member functions are generally applicable matrix functions, they should move to Core/Matrix.

Definition at line 45 of file Classifier.h.

---

The documentation for this class was generated from the following file:

• Core/Tracking/Classifier.h

---